RAWLS’S CONSTITUTION-CENTERED PROPOSITIONS ONLEGITIMACY: A FRIENDLY INTERROGATION

Frank I. MichelmanAugust, 2017

draft of work in progress

General note for NYU readers

The material that follows pulls together doubts I have been nursing about John Rawls’sseeming placement of constitutional law at the load-bearing center of his proposed “liberalprinciple of legitimacy.” Rawls meant by “legitimacy” in politics a morally valued condition inwhich citizens, reasonably perceiving a need, for everyone’s sake, for a high average level ofexpected compliance with law by everyone, can rightfully demand of each other that they dotheir parts toward sustaining this crucial bond of social cooperation.

But, how, then, to justify to supposedly free and equal dissenters the coercive effects oflaws imposed by democratic majorities? Rawls’s principle of legitimacy responds as follows:

Political power is legitimate only when it is exercised in accordance with a constitution (writtenor unwritten) the essentials of which all citizens, as reasonable and rational, can endorse in thelight of their common human reason.

In the work from which this draft is drawn, I follow Rawls sympathetically a good part ofthe way. Accepting the basic outlook of political liberalism, I see his question as aptly posed.I accept also the proceduralistic form of Rawls’s proposed solution—that is, its deflection ofdisagreement over ground-level political outcomes to a hoped-for wide agreement on (or wideagreeability of) an overarching political system or practice from which the potentially divisiveoutcomes issue.

My doubts have specifically to do with Rawls’s seeming reduction, for this purpose, of the“system” (the procedure) to a body of constitutional law in force and institutional arrangementsfor securing compliance. (“Seeming reduction,” because writings of Rawls postdating hispostulations of the constitution-centered legitimacy principle may be read as a move away fromany such reduction. This work of mine may accordingly be construed as an offering of somereasons why.) I believe we cannot hope to reduce a legitimation-bearing political practice to anyset of semi-abstract scriptural norms whose institutionally decisive applications inevitably willreinscribe the disagreements the abstractions are meant to overcome. Constitutional-legal billsof rights and constitutional judiciaries to decide their authoritative applications may (or may not)be parts of a political practice we can judge legitimation-bearing overall. What they cannot beis the whole or the end of it. Systemic justification cannot be teetering at the brink whenever asupreme court erroneously applies a constitution’s scheme of guaranteed rights and liberties, ordoes so in ways that leave many citizens reasonably doubting that the scheme as thus appliedcould be part of a constitution that reasonable and rational citizens have reason to endorse.

I hope the ramble that follows below may adequately clarify whatever in what I’ve just saidmay be murky or puzzling. For this particular occasion, I have thought it fitting to let the ramblesteer into some vintage Dworkiniana and also to start it off by posing an apparent issue betweenRawls and Lawrence Sager, about constitutional law’s receptivity to norms imposing on the statea responsibility for the active pursuit of the conditions of social justice.

I have drawn passages in what follows from a few recent and pending publications, not allof which are cited here. You will have no trouble seeing that this is all very much (as we liketo say) a “work in progress.”

1

I. PRELUDE: JUDICIALLY UNDERENFORCED CONSTITUTIONAL NORMS?

1. Official word from the American Supreme Court—echoed also in many a

jurisconsultative text and treatise—pronounces our corpus juris constitutionalis void of any

norm obliging the state to bestir itself actively in the pursuit of social justice.1 Lawrence Sager

says, though, that if you watch what the Supreme Court does (not just what it says), you will see

clear traces there of the presence in our constitutional law of a “durable . . . commitment to

social justice.”2 Now, when Sager says “commitment,” he means not just some bit of preambular

honking3 but an actual obligation of law, owed by each and every branch of our constituted

governmental system—the Congress, the Executive, the Supreme Court, their counterpart bodies

in the states—to make prevention and correction of “structural injustice” a constant, leading aim

in their respective daily operations.4

2. Sager differs interestingly from UT-Austin colleagues who, rather than insisting on its

presence, instead lament what they see as a recent evacuation (“forgetting”) from our law of any

such obligation.5 That obligation exists there now, Sager claims, as an “unacknowledged” but

still detectably operative norm of American constitutional law. Sager’s resulting complaint is

not that the norm has gone missing, but is rather (implicitly) that relevant actors do not

sufficiently admit to its presence and govern their conduct accordingly.

3. Stuff of law though it be, Sager’s constitutional social-justice obligation breaks the

Hohfeldian mold. It amounts, in effect, to a continuing demand on every American governmental

body to embed its every consequential exertion—from charting agendas and general orders, to

resolutions on guiding principles or policies, to specific acts of legislation and

1 See, e.g., Deshaney v. Dept of Soc. Serv., 489 U.S. 189, 196 (1989) (“Our cases have recognized that the DueProcess Clauses generally confer no affirmative right to governmental aid, even where such aid may be necessary tosecure life, liberty, or property interests of which the government itself may not deprive the individual. . . . Althoughthe liberty protected by the Due Process Clause affords protection against unwarranted government interference, . . .it does not confer an entitlement to such [governmental aid] as may be necessary to realize all the advantages of thatfreedom.”).

2 Lawrence Sager, “The Unacknowledged Constitution” 7 (2017), available athttps://docs.google.com/document/d/1S8jYqI210aIApk7AzHtuVkbBOoR-r4Yg8I0E9nx0cOo/edit?usp=sharing.Professor Sager has this paper in circulation as a draft for discussion, and has kindly agreed to my use of it here.

3 See U.S. CONST. preamble (“. . . in order to establish justice . . .”).

4 Sager, supra, at 8-10.

5 See Joseph Fishkin & William E. Forbath, Wealth, Commonwealth, and the Constitution of Opportunity, inNOMOS LVIII: WEALTH 45, 53-57(Jack Knight & Melissa Schwartzberg, eds., 2017) (describing, explaining, anddeploring a mid-twentieth century “great forgetting” of a formerly vibrant American “language of constitutional politicaleconomy”).

2

administration—in conscientious consideration of the social-justice impacts of choices made and

options forgone. Such an obligation is not of the type giving rise to individual constitutional

rights, at least not directly. Its “correlative” on the demand side of the normative equation (as

Hohfeld might have it) is not anyone’s entitlement or claim-right (or privilege or immunity or

power), but rather a social ideal of normatively constrained political deliberation or public

reason.6

That leaves an obvious question of how the judicial branch fits into the picture, to which

Sager is fully alert. There is “of necessity,” he says,

a division of labor here. Courts address explicit and direct denials of specific rights; legislatureshave to do the work of addressing structural injustice, because the obligation is too porous forthe judiciary to fulfill: What targets are best and appropriate? What machinery will beefficacious? What level or levels of government should assume what part of the burden?7

Sager’s constitutional social-justice commitment accordingly comes complete with instruction

that our courts are not, for its sake, to be tasked with, or on their own to venture into, forms of

judicial oversight and action out of keeping with standard American notions (“justiciability,”

“remedies”) of a bounded judicial power.8 But still he says the commitment is an observably

operative norm in the corpus juris to which our judges show themselves beholden. He makes

an intriguing case, of which a sample comes soon below.

4. Sager’s putative constitutional-legal commitment is to the prevention and cure of

“structural injustice.” And what is that? Partly, at least, it is a cultural condition: in Sager’s

words, the persistence in our social life of “patterns of diminished membership . . . within which

some persons [defined by race, gender, religion and sexual orientation] are systematically

regarded and treated as less worthy by many other members of the community.”9 What about

economy-based, entrenched stratifications of class, which others on the American liberal left

would have us also learn (or relearn) to see as offenses to the Constitution?10 On the evidence

6 John Rawls, The Idea of Public Reason Revisited, 64 U. CHI. L. REV. 765, 767-69 (1997) (hereinafter cited asRawls, Public Reason) (defining an “ideal of public reason” that officials and citizens satisfy by supplying justice-regarding or social-regarding explanations for support of political positions affecting “constitutional essentials andmatters of basic justice”).

7 Sager supra n. 2, at 9.

8 See id. at 8-9.

9 Id. at 14.

10 See Fishkin & Forbath, supra n. 5, at 46-47 (recalling a history of American constitutional debates overquestions of oligarchy, class, and opportunity).

3

of prior writing, Sager would concur.11 His American constitutional social-justice commitment,

then, would correspond in substance with a standard of rightness for a liberal society’s basic-

structural arrangements that John Rawls names as “background justice.”12

5. I do not mention Rawls idly. Had Sager been marching squarely in the footsteps of

Rawls, he might have thought to characterize the social-justice commitment as not, in the main,

a norm of law at all, but rather a norm of American political public reason, by which lawmakers

and administrators are held to account by forces in a working democracy operating mainly

outside of courts (although also occasionally in them). That will not do for Sager, though. His

project is Copernican, Newtonian. His aim is to show how the thesis of the commitment’s

presence now in the material form of law can explain some persisting apparent anomalies in the

practice of a Court for which law provides the sole confessable ground for action.

6. One example will suffice for our purposes. We see the Supreme Court standing ready

to bar as unconstitutional a state legislature’s repeal of an antidiscrimination law (say,

prohibiting private refusals to hire on grounds of sexual orientation) that the Court would have

deemed itself quite powerless to order enacted in the first place13—no doubt in part for the very

reasons noticed by Sager.14 We struggle to explain this seeming anomaly (“How does a return

to . . . a constitutionally benign state of affairs turn into a violation of the Equal Protection

Clause?”15). But if you just think of a Court acting in response to the Constitution’s invisibly

inked social-justice norm—including that norm’s confirmation of the limits of the judicial

power—the mystery evaporates. Viewed (as Sager explains) against the backdrop of that norm,

state public accommodation [and other antidiscrimination] laws are properly understood to belegislative judgments in response to a crucial constitutional obligation to undo structuralinjustice. They are of the same constitutional material as judicial decisions undertaking toenforce the Constitution,

and so a court does not act extrajudicially merely by negating their peremptory withdrawal.16

11 See LAWRENCE G. SAGER, JUSTICE IN PLAIN CLOTHES: A THEORY OF AMERICAN CONSTITUTIONAL PRACTICE

(2004).

12 JOHN RAWLS, POLITICAL LIBERALISM 265-69 (paper ed. 1996) (hereinafter cited as RAWLS, LIBERALISM) (“Therole of the institutions that belong to the basic structure is to secure just background conditions against which the actionsof individuals and associations takes place.”).

13 Compare Reitman v. Mulkey, 387 U.S. 369 (1967); Romer v. Evans, 517 U.S. 620 (1996).

14 See above para 3.

15 Sager, supra n. 2, at 3.

16 Id. at 10.

4

7. Now plainly, for the social-justice commitment to be doing such work as that, we will

have to find it sitting at a high enough level in our legal firmament to be always potentially

controlling over acts and policies of even our highest-ranking constituted legislative, executive,

and judicial powers. Sager accordingly locates it in the Constitution. He treats it, in effect, as

an instance of a class he has elsewhere named as “judicially underenforced constitutional

norms.”17

One may nevertheless ask whether the Constitution is the right location for Sager’s putative

higher norm of law. A possible alternative conception is after all easily available. We could see

the social-justice norm as sitting in a legal-hierarchical space outside and above the Constitution.

We could see it then as directed both to the shaping of constitutional law (which might not,

however, for reasons we’ve begun to note, be suited to containing the whole or the crux of it),

and also—by its own direct force and without mediation through constitutional law—to the

shaping of infra-constitutional, “ordinary” law. (Suggestions that the whole of American law is

itself thus beholden to some pre-constitutional American legal-normative deposit are not exactly

foreign to our tradition.18) The immediate legal force of the social-justice norm would thus carry

over past the Constitution. It would hit directly—not just mediately, through the

Constitution—on ordinary-level legislative deliberations. It would hit them as law, but judicially

underenforced.

8. Well okay, you might say, that is an available alternative construction, but what

advantage might it have over simply absorbing into the Constitution the legal-obligational side

of the American commitment to the pursuit of social justice? My answer is: Sager’s putative

legal norm works its explanatory magic only through its special combination of a mandated

governmental pursuit with a proviso that questions of compliance will not normally be for courts

of law to decide.

17 See Lawrence Sager, Material Rights, Underenforcement, and the Adjudication Thesis, 90 B.U.L. REV. 579,580 (2020) (“judicially underenforced constitutional rights”); Lawrence Gene Sager, Fair Measure: The Legal Statusof Underenforced Constitutional Norms, 91 HARV. L. REV. 1212 (1978).

18 See, e.g., Thomas C. Grey, Do We Have an Unwritten Constitution, 27 STAN. L. REV. 703, 715-16 (1975)(reporting original understanding that the Constitution does not completely codify the higher-law principles that arelegally binding here); Lee J. Strang, Originalism, the Declaration of Independence, and the Constitution: A UniqueRole in Constitutional Interpretation, 111 PENN. ST. L. REV. 413, 417 (2006) (reporting views of some abolitioniststhat “the Declaration [of Independence] abolished slavery or at least rendered it unconstitutional for the federalgovernment to support slavery”). Strang further documents invocations of the Declaration as “an independent sourceof [legally] binding constitutional principles” by 19th-century suffragists, see id. at 422-23, and by 20th-century right-to-life advocates, see id. at 427, 430-31.

5

I make no general objection (as some others might) to a classification as “law” of a mandate

thus trimmed and shorn of full adjudicative backing.19 The possible problem comes, though,

with a certain account of a special function for—specifically—constitutional law in a liberal

state. The account I have in mind has behind it no lesser an authority than John Rawls

(accompanied also, as I will later on suggest, by Ronald Dworkin20). On that Rawlsian account,

to which I am about to turn for extended examination, it seems that—whatever we might say of

law in general—there can be no place in constitutional law for a norm deemed “crucial” but also

nonjusticiable.21

If and to the extent we thus find Sager and Rawls at odds. I do not say the upshot must be

a critique of Sager. It might rather be a critique (or, as I shall rather suggest, a rereading) of

Rawls.

II. “THE LIBERAL PRINCIPLE OF LEGITIMACY”

The Problem of Political Liberalism

9. I tread now myself in the steps of Rawls. We work from within the public culture of

constitutional democracy. We read that culture through a moral-contractual lens, as liberalism’s

modern historical quest for fair basic terms of social cooperation (read also “just background

conditions”) for populations of citizens conceived as free and equal.22

The puzzle, though, is: How does that quest not end up in self-blockade? Political

liberalism, a recently theorized branch of the historical culture marked by special sensitivity to

the challenge posed to that quest by conditions of cultural and moral diversity, confronts us with

the following problem: Given liberal professions of respect and regard for the equal moral

freedom and responsibility of persons, how can we hope to justify societal demands (themselves

no doubt morally inspired23) for everyone’s accommodation of their projects and activities to

19 See Frank I. Michelman, Socioeconomic rights in constitutional law: Explaining America away, 6 INT’L J. CON.LAW 663, 666-67, 684-85 (2008) (noting while declining to adopt a possible objection of that kind).

20 See below para 30 & n. 55

21 See above para 3.

22 See above para 4 & n. 12. For a nutshell version, see JOHN RAWLS, JUSTICE AS FAIRNESS: A RESTATEMENT 18-19 (Erin Kelly ed., 2001) (hereinafter cited as RAWLS, RESTATEMENT) (connecting a “fundamental” political idea ofpersons as each endowed with “moral powers” to “have, to revise, and rationally to pursue a conception of the good”and to “understand and . . . act from . . . principles of justice” to a fundamental political idea of society as a fair schemeof social cooperation among persons thus endowed).

23 See below para 26.

6

requirements of state law, including demands they may find insupportable (i) on grounds moral,

philosophical, or religious, that are (ii) sincerely, responsibly, and respectably held?

Near the start of Political Liberalism, John Rawls poses the question to which his work

there will be addressed. “How is it possible,” Rawls asks, “that there may exist over time a

stable and just society of free and equal citizens profoundly divided” not just by interests but

by “religious, philosophical, and moral doctrines?” Rawls calls that “the problem of political

liberalism.”24 He then immediately follows with what he sees as the same question “put another

way.” How is it possible, he now inquires, “that deeply opposed though reasonable [religious,

philosophical, and moral] doctrines may live together and all affirm the political conception of

a constitutional regime?”25

Enter the constitution

10. Thus seeking a possible practice of legal ordering that can be both stable and just in a

diverse modern society, we find Rawls advancing by way of solution the idea of a

“constitutional regime.” And then asking again about how democratic majorities can hope to

justify the coercive effects of their laws on dissenters no less presumptively entitled than they

are to respect as free and equal citizens, Rawls proposes as follows:

Citizens legitimately exercise . . . coercive political power over one another . . . only when [thepower] is exercised in accordance with a constitution (written or unwritten) the essentials ofwhich all citizens, as reasonable and rational, can endorse in the light of their common humanreason.26

Rawls calls that “the liberal principle of legitimacy.” “Legitimacy” thus names, for Rawls

(and will for us throughout this paper), a morally valued condition in which citizens, reasonably

perceiving a need, for everyone’s sake, for a high average level of expected compliance with law

by everyone,27 can rightfully demand of each other that they do their parts toward sustaining this

crucial bond of social cooperation.

11. Among most or all readers of this, it will go practically without saying that a general

willingness of citizens at large to comply in normal course with duly issued ordinary

24 RAWLS, LIBERALISM, supra n. 12, at xx (emphasis supplied).

25 Id. (emphasis supplied)

26 The formulation just above is found in RAWLS, RESTATEMENT, supra n. 22, at 41. It differs in some details fromformulations in RAWLS, LIBERALISM, supra n. 12, at 137, 217, but not in any respect with which I will here beconcerned.

27 See JOHN RAWLS, A THEORY OF JUSTICE 211 (rev. ed. 1999) (hereinafter cited as RAWLS, JUSTICE) (presenting“Hobbes’s thesis”) (“It is reasonable to assume that even in a well-ordered society the coercive powers of governmentare necessary for the stability of social cooperation.”).

7

laws—most of the citizens, most of the laws, most of the time—is a socially necessary good. But

of course we also know that groups of us are at any time liable to disagree deeply and intractably

(in real political time) over the rightness and goodness of some of a state’s policies as adopted

and pursued by law. How, then, do we sustain across our populations a justified sense of

assurance that our political regime continues as a whole to be deserving of support, even in spite

of your or my severe disapproval of some of the turns it takes or choices it makes from time to

time?

One answer would be: by a credible establishment in the public space of a widely agreeable

test for the regime’s continued wide acceptability overall, despite recurrent severe

dissatisfactions with some of its substantive operations, here and there in the population. It

would have to be a test that those who accept it can furthermore conscientiously see their way

clear to deeming acceptable to everyone for whose acceptance they have good moral reason to

care—say, everyone deemed “reasonable.” Then each citizen could point to and cite this test to

the others in good conscience—each treating the others as equals in dignity and freedom—as

a basis for reciprocating demands for a general disposition to comply with duly issued laws.

According to the Rawlsian principle of legitimacy, a chief and crucial function of constitutional

law is to serve as such a test for a modern, broadly speaking liberal society.

12. We can see pretty well what “reasonable,” here, must mean. We fall back on a set of

perceptions that everyone supposedly could and should share: a perception, first, of the very

great moral and practical benefits to everyone of having some decent system of law effectively

in force; a perception, second, of the persisting facts of conflicts of interests and value-laden

disagreements that might be tolerably understandable on all sides; and then a perception, third,

of the commanding moral logic of a reciprocity of respect for everyone’s quest—yours no less

than mine, a woman’s no less than a man’s, a Muslim’s no less than a Christian’s, a worker’s

no less than a boss’s, and so on—for a life lived in dignity, according to aims and values that

a person affirms for herself or himself in conditions of freedom.

13. With that set of perceptions on board, we then posit the possibility of a set of

constitutional laws for the country that meets the following condition: Each citizen can look the

others in the eye and say, everyone here who shares it should be able to see also that a system

constituted by just these basic laws—here pointing not only to the constitution’s political-

structural arrangements but also its bill of substantive rights—is sufficiently worth upholding

to give each of us prevailing reason to insist on each other’s acceptance (in civic practice, not

in ultimate moral judgment) of whatever issues from the system. So when someone takes

exception to a given policy to be carried out by law, we can feel ourselves morally entitled to

8

respond that the law or policy in question might be right or it might be wrong, it might be just

or it might be unjust, but it is not outside the constitution and so it is in good moral order for us

to call on you for compliance with it.

14. Now, obviously, not just any constitution that may happen to be in force in a given

country at a given time can be allowed to shoulder, in that way, the burden of justification of the

force of ordinary law. It will have to be what we can call a legitimation-worthy constitution

(compare: a sea-worthy ship)—meaning a constitution possessed of whatever the property is that

we think can qualify for such a service a body of basic or constitutional laws. A Nazi

constitution would not, by the lights of my readers, meet this requirement. In his proposed

principle of legitimacy, John Rawls describes the requisite property in terms of the constitution’s

acceptability for such a service, in view of its essential content, to all reasonable and rational,

free and equal citizens.

Legitimacy as a deflection of procedure. “Legitimation-by-constitution”

15. Rawls’s constitution-centered proposition on legitimacy chimes neatly with the strain

of proceduralism that courses through his political philosophy from A Theory of Justice

onward.28 We have, let us say, a group of persons who prefer or who have no choice but to stick

together rather than go their separate ways. In the philosophy of Rawls, “society” (within a given

territory) names such a group.29 Faced with intractable disagreements about questions that could

force their unity apart (do these laws or these policies merit the respect or rather the contempt

of a right-thinking person?), the members find they still can agree on a deflection to a different

question (are these laws or these policies constitutional?), for which they expect a publicly

certifiable answer to be more readily at hand.

That deflection is what I mean by a procedure. The initial question is what we label the

“substance” of the disagreement. The substitute question then is the “procedure” the group

accepts for getting past it.30 (A group of re-uning friends above all committed to spending the

day together decide by coin-flip between the beach and the ball game.) In the Rawlsian

procedural proposition on legitimacy—which I’ve sometimes a bit flippantly called by the name

28 See RAWLS, JUSTICE, supra n. 27, at 172-74 (“In the case of a constitution . . . [the] best attainable scheme isof one of imperfect procedural justice”).

29 See id. at 109-112 (on “the circumstances of justice”).

30 See Frank I. Michelman, Dilemmas of Belonging: Moral Truth, Human Rights, and Why We Might Not Wanta Representative Judiciary, 47 U.C.L.A. L. REV. 1221, 1234-36 (2000) (explaining this sense of procedure and itsapplication to problems of legal controversy).

9

of “legitimation-by-constitution”31— the constitution serves, in that sense, as the stipulation of

the terms of a procedure.

That does not mean, though, that the terms are themselves all and only about political

processes (we vote and the majority rules) as opposed to political outcomes (abridgements of

the freedom of speech are disallowed). Compliance with certain outcome-constraining norms

can constitute a part of the procedural (deflected-to) substitute for the primary question on which

agreement simply is not to be had. So it is with Rawls. His “constitutional essentials” comprise

not only provisions for “the . . . structure of government and the political process” but also for

“equal basic rights and liberties of citizenship that legislative majorities are to respect.”32

“Constitution” as normal-form positive law in Rawls’s proposition on legitimacy

16. When Rawls says “constitution” in his proposition on legitimacy, he has in view a body

of laws within a country’s positive-legal system and practice. Context makes that clear beyond

a possibility of doubt.33 He means laws that are (i) basic, differing in that respect from some

other laws in the system, but that also, in common with all other proper such laws, are (ii)

scriptural, (iii) ascertainable, and (iv) institutionally settleable (with apologies for that ungainly

coinage).

“Basic” means the laws of the constitution (or at any rate the parts that count as “essentials”

in Rawls’s proposition on legitimacy) set terms of recognition for any and all further legal

operations of the state—including most importantly the issuance of directly operative laws by

the everyday legislative bodies. (In Hartian terms, these laws set the system’s highest-order

31 See Frank I. Michelman, Legitimacy, The Social Turn and Constitutional Review: What Political LiberalismSuggests, 3 CRIT. Q. FOR LEGISLATION & LAW 183, 185 (2015).

32 RAWLS, LIBERALISM, supra n. 12, at 227. See RAWLS, JUSTICE, supra n. 27, at 173 (“The political system . .. would not be a just procedure if it did not incorporate [liberty of conscience, freedom of thought, liberty of the person,and equal political rights].”). Compare RONALD DWORKIN, FREEDOM’S LAW: THE MORAL READING OF THE AMERICAN

CONSTITUTION 24-26 (1996), where Dworkin proposes, as the point of substantive constitutional law, the establishmentof basic-structural conditions for imbuing all citizens with a warranted sense of full “moral membership” in thecollectively self-governing political community, by which each can “treat himself as bound together with others in ajoint effort to resolve [political] question, even when his views lose”).

33 In Political Liberalism, Lecture VI, §§ 2, 5, 6, Rawls, in the following order, lays down the constitution-centered principle of legitimacy; defines the constitutional essentials; remarks that he speaks of “a constitutional regimewith judicial review;” endorses the idea of constitutional democracy as legally “dualist,” sustaining both a “higher lawof the people” and an “ordinary law of legislative bodies;” and calls the supreme court an “institutional device[] toprotect the higher law.”

There is room for argument that Rawls at some point after uttering the liberal principle of legitimacy (in PoliticalLiberalism and Justice as Fairness) moved away from thus tethering the moral justification of the force of state lawto compliance with the terms of a positive-legal constitution, see Silje A. Langvatn, Legitimate, but unjust; just, butillegitimate: Rawls on political legitimacy, 42 PHIL. & SOC. CRIT. 132 (2015). My concern just now is with the principleof legitimacy as uttered in those two works. I will take up below the thought that Rawls may later have come to adifferent view.

10

secondary rules.) “Scriptural” means that everyone can point to one and the same set of words,

one and the same collation of canonical prescriptive sentences—fixed (so to speak) in the public

space—and agree that that is what is presently and for some indefinite future to count as is this

country’s set of constitutional essentials. “Ascertainable” means that correct applications of the

essentials “is more or less visible on [their] face” and will only rarely be “open to wide

differences of reasonable opinion.”34 “Institutionally settleable” means that a designated

institutional body’s answers to interpretative questions carry the force of law unless and until

duly institutionally revised.

These elements—of the objectivity, fixity, ascertainability, and institutional resolvability

of constitutional norms—would seem to be already implied by the idea of these norms as law

within a positive legal order. What is more to our point here, though, is how—as a moment’s

reflection will confirm—they seem to be implied by the very idea that constitutionality is to

serve as a procedural test of legitimacy.

III. RAWLS VERSUS SAGER (?)

Rawls’s resulting exclusion of full background justice from legitimation-bearing constitutional

law

17. A procedural deflection for getting past ground-level divisions must have as its premise

that questions of compliance with the procedure are an order of magnitude less open to

reasonable public disagreement than are the ground-level questions themselves, or at any rate

are that much more amenable to trusted institutional resolution. Primarily for that reason, Rawls

excludes anything like Sager’s commitment to a stand against structural injustice from load-

bearing work in his constitution-centered proposition on legitimacy.

“Principles [of justice] covering the basic freedoms and those covering social and economic

inequalities” play different but roles, writes Rawls, in the basic structure of a just society. The

role of the former is to “specif[y] and secure[] citizens’ equal basic liberties and institute[] just

political procedures,” while the role of the latter is to “set[] up the background institutions of

social and economic justice appropriate to citizens as free and equal.”35 The roles are

“complementary,” Rawls says, but they differ decisively in ascertainability:

Whether the . . . essentials covering the basic freedoms are satisfied is visible on the face ofconstitutional arrangements and how these can be seen to work in practice. But whether the

34 RAWLS, LIBERALISM, supra n. 12, at 229 (explaining why a general obligation of pursuit of social justice is nota constitutional essential); see below para 18.

35 RAWLS, LIBERALISM, supra n. 27, at 229.

11

aims of the principles covering social and economic inequalities are realized is far more difficultto ascertain. These matters are nearly always open to wide differences of reasonable opinion;they rest on complicated inferences and intuitive judgments that require us to assess complexsocial and economic information about topics poorly understood.36 Thus . . . we can expectmore agreement about whether the principles for social and economic justice are realized.37

And “these considerations,” Rawls then concludes, “explain why freedom of movement and free

choice of occupation and a social minimum covering citizens’ basic needs count as

constitutional essentials,” while a wider principle of the pursuit of background justice does not.38

For completeness, I must note that “these considerations” include, for Rawls, the additional

thought that, in the pursuit of legitimacy, constitutional-legal assurance in regard to the basic

freedoms is “more urgent” than in regard to the social and economic conditions of background

justice. Why so? Because (so runs the thought) “so long as there is firm agreement on [the

former] and . . . political procedures are reasonably regarded as fair, willing

cooperation”—including in the continuing quest for the conditions of background justice—“can

normally be maintained.”39 I nevertheless say that perceived differences of ascertainability are

the primary and decisive consideration for where Rawls draws the line on the essentials of a

legitimation-worthy constitution, because without those difference there would be no reason to

exclude any guarantee that Rawls would regard as truly and fully a principle of justice for a

liberal society.

18. But why, you might ask, should difficulties of ascertainment of compliance with a

guarantee exclude that guarantee from the terms of a constitutional-legal test for legitimacy,

when that test stands to be applied by a trusted institutional decider? Granted, potentially

divisive disagreements over ground-level laws and policies cannot be overcome by a resort to

abstractly stated “constitutional essentials” on which all can agree, but only because they paper

over the persisting disagreements that inevitably will surface at the point of application of those

essentials to concrete policies and actions. It does not follow, though, that citizens must always

agree in the first person (so to speak) about the finally correct answers to questions of

fulfillment of the constitutional essentials. Citizens might instead find it reasonable to defer to

an institutional service whose considered judgments regarding such questions are widely trusted

to fall within the bounds of honest, discursive defensibility—not, of course, infallibly but with

36 Compare Sager’s comments on “division of labor,” above para 3.

37 RAWLS, LIBERALISM, supra n. 12, at 229-30.

38 Id.

39 Id. at 230. Not all liberal-minded thinkers concur. See TOMMIE SHELBY, DARK GHETTOS: INJUSTICE, DISSENT

AND REFORM (2016).

12

a frequency sufficient to qualify those judgments as publicly authoritative for legitimacy-

sustaining purposes. And indeed it is precisely with a view to this crucial moral function—to

assist in the enablement of political legitimacy on liberal terms—that Rawls himself defends the

use of courts as authoritative public arbiters of the fulfillment of the constitutional essentials.

But from that very conception of the Supreme Court’s service “as the highest judicial

interpreter of the constitution” flows a limit on the allowable scope of reasonable debatability

of questions of compliance with the legitimacy-bearing constitutional essentials. For that service

requires a supposition of public confidence that

the political conceptions judges hold and their views of the constitutional essentials locate thecentral range of the basic freedoms in more or less the same place. In these cases at least itsdecisions succeed in settling the most fundamental political questions.40

All told, then, we have to understand Rawls as concluding that a constitutional guarantee of

sufficient dedication in the pursuit of the social and economic conditions of background justice

cannot stand up to that prerequisite for full public trust in the Supreme Court.41

19. This is a result that many, including Lawrence Sager, will find hard to take—some of

us for reasons we learn from Rawls. In order to meet the Rawlsian test for legitimation-

worthiness, a political order should include a commitment affecting every topic for which a

rational and reasonable person would reasonably seek one as a fair condition for willing support

for the system as a whole. Now, can we reasonably call on everyone as free and equal to submit

their fates to the tender mercies of a democratic-majoritarian lawmaking system, without also

committing our society credibly, from the start, to run itself in ways designed to constitute and

sustain every person as a competent and respected contributor to political exchange and

contestation as well as to social and economic life at large? If not, then a commitment to the

pursuit of background justice should be salient in the publicly acknowledged basic terms of the

lawmaking system with whose outputs everyone is called upon to comply. Any expression of

the morally requisite systemic commitment will be justifiably suspect if we begrudge it full

expression in the social form that begets the maximum civic backing the country's political

40 Id. at 237.

41 I have dealt with this aspect more expansively elsewhere. See Michelman, supra n. 19, at 683-84 (explaininga widely held concern that “writing or reading socioeconomic assurances into constitutional law would run risks ofserious damage to the integrity (and to public confidence therein) of the country’s practices of constutionalism[and]legality . . . upon which political legitimacy depends”).

13

culture is able to muster. It would seem to be a premise for Sager and others that that form, in

our country now, for better or for worse, is constitutional law.42

B. Reconciliation?

20. It looks as though that brings Sager up against Rawls, at least to some extent. But to

what extent, exactly? The two agree that a commitment to the pursuit of background justice is

a requirement for a society aspiring to a fully just condition. They agree on a need, even so (and

also for the sake of justice), to defer to a perception of nonjusticiability-at-the-core of a guaranty

of sufficient dedication in the pursuit of background justice. The difference is only that Sager

points out how the guarantee is still justiciable around the edges. So why not (we might imagine

Larry asking Jack) include it as an essential for any liberally legitimation-worthy constitution,

just up to and not beyond the edge of nonjusticiability?

That will not do. If public confidence in the adjudicative calling out and correction of

breaches of essentials itself is to count (para 18) as a kind of secondary procedural prerequisite

of legitimacy, then either sufficiency of dedication would have to be justiciable at the core (not

just around the edges), or else it cannot be a Rawlsian constitutional essential.

But Rawls and Sager could still agree, could they not, on including sufficiency of

dedication as a judicially underenforced part of constitutional law, only not counting it as a part

on which legitimacy depends? And isn’t that, then, the respect in which we might see Sager as

offering a friendly improvement to Rawls?

Well, that is a question to Sager, not to me. He speaks of a “crucial” constitutional

obligation to undo structural injustice.43 But crucial to what? is now the question. Crucial, no

doubt, to the pursuit of justice full and true. But does Sager mean also crucial (even) to

legitimacy, the condition (as Rawls puts it) in which “willing compliance can normally [but

Rawls here also means “justly,” in an extended sense of that term] be maintained.” If so, then

Sager looks to be on a collision course with Rawls.

But I take these two to be friends who would rather not run each other over, so it looks like

one of them will have to veer. I count myself a friend to both, and I am going to suggest it should

be Rawls who veers. It is not only Sager’s implicit challenge that so inclines me. Sager is

figuring for me here as the canary in the Rawlsian depths, his chirp a signal that something in

the air may be amiss.

42 The same view is also very strongly manifest in the work of Fishkin & Forbath, supra n. 5.

43 See above para 6.

14

IV. SOME CONSEQUENCES OF LEGITIMATION “BY CONSTITUTION”

21. Justification of divisive ground-level political acts by appeal to the legitimation-

worthiness of an overall political system or practice from which they issue is what we may call

an idea of “legitimation by system.” “Legitimation by constitution” then stands for a variation

on that theme, by which “the system” that is to bear the justificatory load—the system, that is,

on the perceived legitimation-worthiness of which will depend the moral justification for a

reciprocity of demands for respectful submission to laws issuing from it—is reduced to a body

of positive-legal material, the “constitutional essentials.” We have just been examining into one

consequence of such a reduction, to wit, a resulting resistance against writing into constitutional

law an active state obligation to the prevention and cure of structural injustice. In the sections

that follow below, I take up some other apparent consequences.

A duality of aims for substantive constitutional law

22. A first and general consequence, affecting others to follow, is a duality in the kinds of

aims with which societies like ours may invest the substantive parts of their basic-law

constitutions, which I will here name as a “regulatory” and a “justificational” aim. (Of course

this dual classification does not in itself conjure up the full and rich array of reasons and motives

that theorists and observers may find for the introduction into a country’s legal practice of a

layer of substantive constitutional law.44 The classification only sorts out these sundry possible

aims along one axis of differentiation among them.)

The regulatory function of a constitutional bill of rights is to constrain political outcomes

over a span of time in directions preselected by the authors (whom we may or may not idealize

as “the people”). The justificational function is to provide for everyone concerned a good and

sufficient reason right now, in the present moment, for willing submission to the laws that issue

from everyday lawmaking, which may run strongly against the interests or beliefs, including

moral beliefs deeply held, of substantial numbers of citizens.

23. The “liberal principle of legitimacy” propounded by Rawls assigns a justificational

function to substantive constitutional law. Yet it seems that whoever makes that assignment

must be assigning a regulatory function, too. Justificational force presupposes regulatory effect.

Justification says (roughly speaking): “Yes, we have disagreements, but still we all can (and

44 See, e.g., RAN HIRSCHL, TOWARDS JURISTOCRACY: THE ORIGINS AND CONSEQUENCES OF THE NEW

CONSTITUTIONALISM 43-47 (2004) (advancing a “hegemonic preservation” thesis to explain constitutionalization ofsubstantive rights); Dieter Grimm, Integration by Constitution, 3 INT’L J. CONST. L. 193, 195 (2005) (advancing thethesis of an expressively “integrative” function for substantive constitutional law).

15

therefore should) go along with the legislative outputs of the legal order in force—given that

(which is to say, just insofar as) we have assurance that those outputs do in fact issue in

conformity to certain instructions pre-inscribed in this constitution (to which we are just now

pointing).”

From this, it may seem we can expect that the regulatory and justificational functions of

a substantive constitution will coincide, so that pursuit of latter must also involve a pursuit of

the former. On closer look, though, one finds that by loading the justificational (or legitimation)

function on top of the regulatory, we set up puzzles of execution of constitutional

responsibilities by judges, lawmakers, and citizens.

A puzzle of citizen responsibility. “Civility” as political obligation

24. Suppose a constitution has clauses on “liberty,” “equal protection,” “free exercise” and

“non-establishment” of religion. The country debates whether that combination (plus Preamble)

adds up to (i) a constitutional ban against exclusion of same-sex couples from state-sanctioned

marriage plus (ii) putting elected county clerks to a choice between keeping their jobs and

conscientiously refusing to make themselves party (by official signature) to the facilitation of

gay marriages. At the moment, no competent authority having spoken to the contrary, these

matters remain within the keeping of the ordinary legislative bodies to decide in whatever ways

they may choose. Next week, though (let’s say), the country will be voting on a proposed

constitutional amendment that would specifically prohibit any law or other official action

conditioning access to marriage on the genders of the partners. You as citizen have a vote to

cast. And you, as it happens, subscribe to a liberal moral conception that treats as falling within

the scope of fundamental human rights a person’s freedom of choice of a marital partner

regardless of sex. Might the idea of a justification-bearing constitution nevertheless point you

toward leaving open for some while longer the question of the permissibility of legislative

restrictions on same-sex marriage—and thus toward a “no” vote from you on your country’s

pending constitutional amendment?

25. A constitution’s justificational function succeeds, when it does, as a procedural

deflection from ground-level issues subject to reasonable but potentially divisive disagreement.

Such a deflection, we have noticed,45 must have as its premise that questions of compliance with

the procedure are an order of magnitude less open to reasonable public disagreement than are

the ground-level questions themselves, or at any rate are that much more amenable to trusted

45 See above para 17.

16

institutional resolution. Any substantive content in the procedure must accordingly be kept

sufficiently thin or abstract to avoid foreclosures of matters of grave moral moment on which

agreement does not yet exist throughout the population of reasonable citizens. To that end, in

the view of Rawls, the list of substantive essentials for a legitimation-worthy constitution is to

be short and its items cast at accommodating levels of abstraction.46 The strategy is of one of

postponement of unity-breaking disagreement to beyond the moment of assessment of the legal

constitution’s legitimation-worthiness.

Thus, while a legitimation-bearing constitution may have to say something in the way of

substantive guarantees, it cannot possibly, at any point in time, prescribe for everything that

might indeed be a true requirement of human right—so long, that is, as you accept also that what

is or is not a true requirement is sometimes open to reasonable disagreement. Even with such

a seemingly obvious matter as liberty of conscience, there will be limits on what the constitution

can say and in how much concrete detail. The boundaries will vary and shift from country to

country and from time to time.

How, if at all, ought these considerations bear on how you will cast your vote on the gay-

marriage amendment?

26. To that perhaps not-so-welcome question you might think of this possible retort: The

reason the rhetorical question insinuates for a possible “no” vote on the same-sex marriage

amendment is a reason, you might say, of expediency—that is, to accommodate demands

coming from others whose cooperation you urgently need—and not of rightness. And

considerations of expediency cannot affect your vote where rightness, in your considered

judgment, points clearly in favor of “yes.”

A follower of Rawls would have, I think, a two-step response. First, what is up for decision

here is not the state of the everyday law that immediately governs the question of gay marriage.

Rather it is the state of basic law, the law that sets the terms for valid lawmaking. Second, when

it comes to basic laws, the relevant considerations of rightness and expediency do not line up

as simple opposites in the way your retort supposes.

When voting on a constitutional amendment, what is at stake, if you follow the Rawlsian

argument, is something different—some might say it is something “larger”— than gay-marriage

46 See RAWLS, LIBERALISM, supra n. 12, at 232 (“ The principled expression of higher law is to be widelysupported,” and so “it is best not to burden it with many details and qualifications.”); id. at 296 (“Whenever we enlargethe list of basic liberties we risk . . . recreating within the scheme of liberties the indeterminate and unguided balancingproblem we had hoped to avoid by a suitably circumscribed notion of priority.”).

17

policy or free-speech (or whatever) policy. It is the legitimation-worthiness of the resulting

constitution. And when it comes to catering for a constitution’s legitimation-worthiness,

prudential and moral considerations are not opposable in the way your retort takes for granted.

That is because a failure of political justification at the system level—or so it may be

contended—is not a merely prudential or pragmatic mishap but a moral mishap as well.

Why so? Because such a failure strips a country’s citizens of recourse to reasons that

everyone should find acceptable for a mutuality of expectations of a prevailing regularity of

compliance with that country’s laws by everyone. It seems that should not be a morally tolerable

outcome for a people who claim to prize each person’s free development and exercise of his or

her moral and other capacities, and on that very ground to find a moral necessity both in the

support of civil government and the force of legitimate law and in the creation of (in Rawls’s

words) “a social world in which [all], as free and equal, can cooperate with others on terms all

can accept.”47 It seems a pursuit of that latter condition must be a part what we owe to each other

as a matter of basic recognition and respect. And then that, too—that moral duty of civility, as

John Rawls calls it,48 that obligation of respect toward disagreeing fellow citizens as

presumptively free and equal, reasonable and rational—must itself qualify as a part of the

background morality against which constitutions are to be rated as morally better and worse.

Attribution to basic laws of the justificational function thus goes hand in hand with the idea

of a moral obligation of civility. Of course the cluster of values here coming to the fore—moral

powers, recognition, fair terms, reciprocity, civility, reasonability—are of the sort that John

Rawls would classify as “political,” not “metaphysical” or “comprehensive.” But this

classification does not make them less than full-fledged moral values. It rather makes them

moral values of a special kind (Rawls speaks of the “the very great values” of the

political”49)—and indeed of a kind that takes on, in a Rawlsian view, a leading and in some ways

a controlling role in any conception of justice for a modern pluralistic state.

Like it or not, that is what political-liberal reliance on the constitution for a justificational

service seemingly must come to. From which it inexorably follows, disturbing as the thought

may be, that, for any country in conditions of reasonable pluralism, its constitution almost

certainly cannot rightly, at any point in time, prescribe directly for everything that is indeed

(when the right answer is finally at hand) a true requirement of human right.

47 RAWLS, LIBERALISM, supra n. 12, at 50.

48 See id. at 217-18, 226, 236, 242, 253; RAWLS, RESTATEMENT, supra n. 22, at 90, 92, 118.

49 RAWLS, LIBERALISM, supra n. 12, at 139.

18

A puzzle of judicial constitutional interpretation

27. If (para 25) the strategy of legitimation-by-constitution is of one of postponement of

unity-breaking disagreement to beyond the crucial moment of assessment of the legal

constitution’s legitimation-worthiness, the price is that the resulting thin or abstract body of

substantive constitutional law cannot be self-applying. Its applications must often be subject to

reasonable debate, and that is why we need the Supreme Court. So suppose, now, the case posed

in para 24, except that the questions come as matters of application by the Court of the

constitution as it stands, sans amendment. The county clerk has turned away the same-sex

couple; the couple have sued for relief on constitutional grounds, and the matter is now before

the Supreme Court to decide.

Consider now this pairing of two statements by Rawls:

(1) “[C]onstitutional democracy is dualist: It distinguishes . . . the higher law of the people fromthe ordinary law of legislative bodies. . . . A supreme court fits into the idea of dualistconstitutional democracy as one of the institutional devices to protect the higher law . . . [aslong as] its decisions reasonably accord with the constitution” as laid down by the people.50

(2) “The justices [interpreting the constitution] . . . must appeal to the political values they thinkbelong to the most reasonable understanding of the public conception . . . . These are valuesthey believe in good faith . . . that all citizens as reasonable and rational might reasonably beexpected to endorse.”51

These two remarks appear to sort nicely with our functions of regulation and justification (para

22); they do not, however, seem to be pointing in the same direction.

28. Where constitutional law is to serve as the medium of fixation by the authors (as we

say, the people) of certain general aims regarding future political outcomes, the corresponding

assignment to judicial appliers must be, as Rawls says in the first of our paired remarks, to

“protect” the work of the authors. We must then expect from these appliers their best effort at

extracting from the words and surrounding facts the historically enacted will of the authors.

That authorial will may of course move on a moderately abstract plane of principle. Say,

it would be a will to condition the validity of any later-arriving law on its due deference to a

principle envisaged and named by the authors as “the freedom of speech”—and so further on

that later law’s deference to whatever that principle—their principle—may turn out to

encompass in future applications not expressly considered by them, in social conditions perhaps

not foreseen by them. The task, though, must still remain one of historical-factual inquiry into

what the authors envisaged as the gist and content of the principle thus named.

50 Id. at 233-34.

51 Id. at 236.

19

Now, we know from experience that inquiries of that kind will quite frequently run out

before they have turned up decisive answers either way to current controversies over the

political-moral merits of various legal acts. Where they do, Rawls’s court-as-protective-device

proposition does not, just standing on its own, give the judges in a democracy a foothold for

ruling against the choices (packed with their own express or implied constitutional

interpretations) of the state’s executive and legislative authorities. Or so the case has appeared

to many.52

29. By seeming contrast, where the constraints of constitutional law are to supply sufficient

justification now for willing submission by dissenters to the coercions of ordinary law—the

second of our paired statements by Rawls— the corresponding assignment to judicial appliers

must be (within some outer limit of current semantic defensibility) to enforce a set of

constitutional essentials that really does supply the needed justification. As to what “really

does” supply it, though, the judges will be at least in some degree on their own, as no one has

better shown than Ronald Dworkin.

30. As persuasively explained by Dworkin, any truly respectful applier of a constitutional

text will read the words against a backdrop of the authors’ own supposed conception of the

political-moral “upshot or point” of writing a basic-law constitution in the first place. A political

community’s commitment to such a document with its particular provisions necessarily reflects

their “prior commitment to certain principles of political justice which, if we are to act

responsibly, must therefore be reflected” in the way we now read the Constitution. A reader,

Dworkin says, cannot truly show regard for either a constitutional text or “the motives of those

who made it” without ascribing to them certain “principles of political morality which in some

way represent the upshot or point of constitutional practice more broadly conceived.”53

Yet it seems also that any conception of “upshot or point” ascribed to authors by a

respectful interpreter will have to be one that makes good sense to the interpreter. The

interpreter “proposes value for the practice by ascribing some scheme of interests or goals or

principles the practice can be said to serve or exemplify.” But plausibly defensible ascriptions

will differ among interpreters, so any interpreter’s choice will have to reflect his or her own

52 See, e.g., Robert H. Bork, Neutral Principles and Some First Amendment Problems, 47 IND. L.J. 1 (1971).

53 RONALD DWORKIN, A MATTER OF PRINCIPLE 35-36 (1985). Dworkin soon thereafter would generalize the pointbeyond its application to “constitutional practice” to the larger social practice known as “law.” See RONALD DWORKIN,LAW’S EMPIRE 87 (1986) (“Law is an interpretive concept. . . . Judges normally recognize a duty to continue . . . thepractice they have joined.”); id at 66 (“[T]here must be an interpretive stage at which the interpreter settles on somegeneral justification for the practice . . . . This will consist of an argument about why a practice of that general shapeis worth pursuing.”).

20

view of which ascription “proposes the most value for the practice—which one shows it in the

better light, all things considered.”54

Suppose now a political-liberal-minded judge, who finds that the “upshot or point” of

constitutionalization of substantive norms is to ensure justification—not just on the day the text

is written but over some future course of political time, while that constitution remains in

place—for calls among fellow citizens for willing submission to laws with which some of them

may disagree profoundly.55 Then (passing now from Dworkin back to Rawls), today’s judicial

appliers must read and apply the words (again, the second of our paired remarks of Rawls on

constitutional interpretation) in the light of “political values . . . that they [the judges] believe,

in good faith . . . all citizens as reasonable and rational might reasonably be expected to

endorse.”56 The legal text will not be disregarded, but it will be read against the backdrop of a

political-moral purpose that the judicial reader cannot simply find already in the text (because

it indispensably informs the reading of the text) and so must of necessity bring to it.

31. It seems, then, that between the regulatory and the justificational modes of

constitutional application there must always remain some gap of difference in the questions

presented to the applier. Alessandro Ferrara has put it crisply: “In . . . the first mode the

interpreter of the constitution is asked to tell the public what the sovereign people did will, in

the second to tell us what it should will.”57 If we are lucky with our historical contingencies—if

our finding of what the people did will matches our view of what they should will—that gap will

not matter in practice. Apply the constitution, then, for the sake of regulation in accordance with

the authors’ directions, and you will also ipso facto apply it with regime-justifying effect.

A classic case, that, of a Levinsonian happy ending!58 It follows where the constitutional

authors happen to have constitutionalized all the principles whose observance is required to

make a democratically and liberally justifiable regime—and none that would defeat it. And what

54 Id. at 52-53

55 Compare RONALD DWORKIN, FREEDOM’S LAW: THE MORAL READING OF THE AMERICAN CONSTITUTION 24-26(1996), where Dworkin, for his own part, proposes as the point of substantive constitutional law the establishment ofbasic-structural conditions for imbuing all citizens with a warranted sense of full “moral membership” in thecollectively self-governing political community, by which each can “treat himself as bound together with others in ajoint effort to resolve [political] question, even when his views lose.” Dworkin, if I read him right, points straight ata justificational function for substantive constitutional law.

56 RAWLS, LIBERALISM, supra n. 20, at 236 (emphasis supplied).

57 Alessandro Ferrara, Political liberalism revisited: A paradigm for liberal democracy in the 21st century, 24PHIL. & SOC. CRIT 681, 686 (2016).

58 See Colloquy, Fidelity as Integrity, 65 FORDHAM L. REV. 1357, 1358 (1997) (statement of Sanford Levinson).

21

if (in your or my liberal-minded estimation) they have not? Or rather—the more deeply

worrisome question—what if our condition of reasonable pluralism blocks us from agreement

on whether they did, and, if so, to what effect?

Does the combination of “liberty,” “equal protection,” “ free exercise,” and “no

establishment” add up to a ban against exclusion of same-sex couples from state-sanctioned

marriage plus making county clerks choose between keeping their jobs and abstaining from

endorsement of gay marriages? Either answer to Ferrara’s “did will” question still leaves for

answering the question whether the resulting constitution is legitimation-worthy. Either answer

to the “should will” question still leaves for answering the question whether that is what they

did will. The two questions simply and incurably are different. Not only different, independent.

No logic supplies an inference from the answer you give to one to the answer you give to the

other.59

32. Piling the justificational function on top of the regulatory thus leaves interpreters caught

betwixt and between. This result chimes easily with the interminable American debate between

textualist/originalist and ”philosophical” or “moral reading” approaches to judicial

constitutional interpretation.60 (As interestingly, it chimes with an observation that not even our

most avid moral readers can ever quite forswear the idea of a publicly legible fixture in time of

the constitutional essentials.61) We simply read that debate (or that choice) through a

justificational lens (as political liberalism proposes), but construing it as a contest between (i)

59 If prepared to face down another “happy ending” jeer, you might think of suggesting that history itself suppliesthe logical link, because what it shows over the long arc is that what the people in this blessed country of ours did anddo will above all is a truly legitimation-worthy constitution and that judicial and other appliers should interpretaccordingly. Philosophical-minded judges have been known to take that tack. See United Mizrahi Bank Ltd. v. MigdalC o o p . V i l l . , 4 9 ( 4 ) P . D . 2 2 1 ( 1 9 9 5 ) , E n g l i s h t r a n s l a t i o n a thttp://elyon1.court.gov.il/files_eng/93/210/068/z01/93068210.z01.pdf, at 271-72 (Barak, President) (“By means ofjudicial review we are loyal to the fundamental values that we took upon ourselves in the past, that reflect our essencein the present, and that will direct our national development as a society in the future.”)

60 See JAMES E. FLEMING, FIDELITY TO OUR IMPERFECT CONSTITUTION: FOR MORAL READINGS AND AGAINST

ORIGINALISMS 80-81 (2015) (“philosophic” approach); RONALD DWORKIN, FREEDOM’S LAW 1-3 (1996) (“moralreading”).

61 James Fleming describes the philosophic approach as a mode of interpretation in which judges progressively“strive for the true meaning or best account” – of what? Not of what a Rawlsian justification-worthy constitutionhypothetically should say, but rather of what “our” constitution does say – if sometimes only somewhat broadly andabstractly – to wit, “our . . . ends-dedicated scheme of . . . rights phrased more often than not as referring to generalgoods and principles.” The philosophical approach, Fleming says, “does not involve judges or other interpreters in doingmoral and political philosophy without regard to the commitments of our constitutional order.” Id. Or compare thisrestatement of Fleming’s view offered by Jack Balkin: Constitutional words and sentences “bear an implicit moral orpolitical logic, and so constitutional interpreters should read them “as examples or instantiations of a prior or moregeneral moral or political theory that underlies the Constitution.” ”Jack M. Balkin, History, Rights, and the MoralReading, 96 B.U. L. REV. 1425, 1434-35 (2016).

22

the formal-democratic view that a true historical fact of effective higher-law enactment by the

people (with whatever in the way of substantive higher law the people may or may not have seen

fit to lay down) suffices to justify the daily force of conforming ordinary law, and (ii) the view

that justification depends on a further, substantive assessment of the legitimation-worthiness of

the higher law in force.

A puzzle of closure versus openness of the constitutional essentials

33. Suppose we join with Rawls in taking up stance (ii). And now please look again at the

paired statements from Rawls in para 27, above. Are we not still caught betwixt and between

the “protect the people’s higher law” view (1) and the “political values they think right” view

(2)?

Justificational force, we said above,62 presupposes regulatory effect. Justification says: “So

long as the legal system in force requires and assures that legislative majorities will uphold these

norms—pointing by “these” to the scriptural constitution, the basic law—the system should in

all reason be found acceptable among free and equal citizens, and so we may all justifiably insist

on each other’s compliance with the laws that duly issue from the system. But that “so long

as”—so the argument will run—sends us right back to the regulatory function of substantive

constitutional law. The county clerk and her sympathizers seek to know now whether the system

is or is not one that gives her assurance that, come next week or next year, she will be able to

send same-sex marriage-seekers to the next window or the next county and still retain her post.

They already have their settled convictions about what a legitimation-worthy constitution would

say about that matter, what they need further to know (in order to know whether the force of

ordinary law is justified in accord with the Rawlsian principle of legitimacy) is what “this”

constitution does say. A legitimation-bearing body of constitution law cannot be a pig in a poke.

34. Okay, a legitimation-bearing body of constitutional law cannot be pig in a poke. And

yet it has to be, by Rawls’s constitution-centered proposition on legitimacy. In order for

adherence to a legitimation-worthy set of constitutional essentials to serve as a procedure for

getting past intractable divisions of substance, it seems those essentials must at all times be

framed in terms sufficiently open to leave many divisive questions of applications presently

unresolved, postponed to some indefinite future resolution.63 That requirement of openness

holds not just during some initial sign-up period, while citizens reason their ways toward

62 In para 23.

63 See above paras 15, 25.

23

agreement that they and others can reasonably stake their freedom and equality on a system

forever bound by the scripted set of constitutional essentials. The requirement of openness holds

continuously throughout a free society’s unfolding future course of generational successions,

social learning, and new conditions disclosing new issues of application or casting old one in

a new light. (This is what Rawls means by positing a “fact” of reasonable pluralism, presenting

a (permanent) “problem” of political liberalism.64) Always either postponed or re-openable must

be excessively divisive applications—we are not talking minor details, here—that you and I and

county-clerk sympathizers will care deeply about. So it must be the case, for Rawls, that at any

moment there will be questions of application for which either a “yes” or a “no” answer will be

“in accordance with” with the constitutional essentials. But then (the question from para 33)

how can the scripted constitutional essentials possibly stand as the country’s public contract (so

to speak) on legitimacy?

V. REREADING RAWLS

35. Rawls now looks like wanting to have it both ways: The constitutional essentials both

cannot be and must be a pig in a poke. That’s so, you might well respond, but only because I

have been reading Rawls to make legitimacy depend on an assurance that legal officials will

reach, at every step, singularly right answers to questions of application of a scripturally

formulated scheme of constitutional essentials. But Rawls never says that. He says legitimacy

subsists as long as coercive political power is exercised “in accordance with” a legitimation-

worthy set of constitutional essentials. Now, who says “in accordance with” has to mean

“without getting applications wrong”?

But if Rawls does not mean that, what else might he mean? He might mean that

applications are conducted by those entrusted to conduct them in a manner that is observably

competent and sincere, open and fair, with due regard for the scripted rights and principles but

also with a controlling eye always to the question of ultimate import: to wit, whether the

resulting basic-law system remains as one that reasonable and rational citizens can find

acceptable in the circumstances (and here please revert to the remarks on “reasonable” in para

12, above).

What then finally matters will be the spirit or ethos with which applications of the

constitutional essentials are conducted, not the getting of them Herculeanly “right.” That some

64 See RAWLS, LIBERALISM, supra n. 12, at xviii-xix, 4, 36-37 (“Political liberalism assumes that . . . a pluralityof reasonable yet incompatible comprehensive doctrines is the normal result of the exercise of human reason within theframework of free institutions of a constitutional democratic regime.”); above para 9.

24

applications that you or I or county-clerk sympathizers may care deeply about remain unresolved

for the moment does not prevent a present favorable judgment of that spirit or ethos, and neither

does the occurrence of a resolution we oppose prevent a favorable judgment on the record as a

whole. The scripturally listed “equal basic rights and liberties of citizenship that legislative

majorities are to protect”—a/k/a “constitutional essentials”—are to serve, then, not as the rigid

terms of a fragile contract but rather as discursive guidelines, settled anchoring points for a

constitutional public reason aimed finally at the protection, not of a legal contract, but of an

overall legitimate political practice according to the standard set by Rawls: acceptable to free

and equal, reasonable and rational citizens.65

36. A perception of need to clarify this point may be a part of what led Rawls to offer, in

a writing that apparently postdates his prior formulations of a “liberal principle of legitimacy”

(the ones I have been working with throughout), a formulation that by its parallel structure

seems meant to replace them. Rawls called it by a new name, “the idea of political legitimacy

based on the criterion of reciprocity.” It says:

Our exercise of political power is proper only when we sincerely believe that the reasons wewould offer for our political actions—were we to state them as government officials—aresufficient, and we also reasonably think that other citizens might also reasonably accept thosereasons. This criterion applies on two levels: one is to the constitutional structure itself, theother is to particular statutes and laws enacted in accordance with that structure. To bereasonable, political conceptions must justify only constitutions that satisfy this principle.66

Such a revisionist reading of Rawls—replacement of legal assurance of outcomes by

assurance of a prevalence of public reason, at the load-bearing position in systemic political

justification—does no doubt run into all the contrary tending remarks and positions that I have

been pointing to above.67 (It leaves Rawls’s case for excluding the pursuit of background justice

from the constitutional essentials gravely weakened if not without a leg to stand on.68) The spirit

of Ronald Dworkin is here to remind us that such interpretative contretemps do not necessarily

65 Compare ALESSANDRO FERRARA, THE DEMOCRATIC HORIZON: HYPERPLURALISM AND THE RENEWAL OF

POLITICAL LIBERALISM 48-51 (2014) (proposing treatment of established liberal “fundamental rights” not as “normativestumbling blocks that political will cannot ignore without forfeiting its legitimacy,” but rather “a moment of closurethat an open-ended political will freely creates for the sake of better preserving the openness of the political process”).

66 Rawls, Public Reason, supra n. 6, at 771. This formulation certainly postdates the ones in RAWLS, LIBERALISM,supra n. 12, and probably also postdate the one in RAWLS, RESTATEMENT, supra n. 22, published in 2001. See id. atxii (editor’s remark that “by 1989 the manuscript had evolved into something close to its present form”). I owe thisobservation to correspondence with Silje Langvatn

67 See above paras17-19, 33-34.

68 See below para 39.

25

make it not the best reading overall. Any Herculean reader of Rawls—yes, even of Rawls—will

have to shoulder the burden of casting off some parts of the record as (ahem) “mistakes.”69

36. So what, then, would be the putative mistake? “Seeming to pin legitimacy too strongly

to full and correct enforcement of a constitutional-legal contract” would approximately describe

the one I am have been after. But still we have to assign some work, some function to those

“constitutional essentials,” and reducing them, as I have suggested to anchoring points for

constitutional public reason, may still leave us with a problem.

Why, afer all, these particular anchoring points? The moral work they do must still be in

service to some conception—some distinct, some fixed conception, thinned out as it may be—of

the non-controvertible basics of government practice here that any legitimation-worthy

constitution, by Rawls’s later reciprocity-based formulation, still must “satisfy.” Silje Langvatn

(to whose work I stand much indebted for the idea of the re-reading) does herself read Rawlsian

political reciprocity as a distinctly liberal-tinctured idea. As judge, as official, as citizen, my

votes on matters of basic justice and constitutional essentials are to be in accord with that I

sincerely hold to be “a reasonable and sufficiently complete interpretation of the basic political-

moral ideas of the public political culture of constitutional democracies,” which include “the

idea of having a constitution that protects a set of liberties for each.” My aim in such cases is

always to be “to interpret, specify, and give institutional effect to the basic political-moral ideas”

of an extant regime supposedly of that type, in the face of disagreements about “how to best

combine and translate [these political-moral ideas] into specific legal and institutional

arrangements.”70

37. Do we thus invite in by the back door the worry this whole story starts out from, about

taking pluralism seriously?71 Alessandro Ferrara has named as a fact of “hyperpluralism” in the

societies we deal with those clashes of faith and culture that seem to thin out to the point of

evaporation any prospect for agreement on a set of substantively liberal-tending constitutional

essentials.72 As a possible response, Ferrara suggests, not an abandonment of the liberal

constitutional essentials, but a further inclusion in an affected country’s basic-structural

69 See RONALD DWORKIN, TAKING RIGHTS SERIOUSLY 118-23 (1977) (on the need of Hercules, J., for a “theoryof mistakes”); cf. Langvatn, supra n. 33, at 133 (“[I]n many of Rawls’ texts ‘superseded thoughts appear to be retainedalong with later ideas’, and in some early texts Rawls hints at ideas that he develops only much later” (quoting GeraldGaus, The Turn to a Political Liberalism, in A COMPANION TO RAWLS 233, __ (J. Mandle & D. Reidy, eds., 2014)).

70 Langvatn, supra n. 33, at 142-43.

71 See para 9, above.

72 See FERRARA, supra n. 65 at 89-91.

26

arrangements of certain additional components to ease the pressures for substantive agreement:

provisions for legislative devolution (federalism, bi-nationalism) or for institutionalized legal

pluralism within a unitary state. In this response, some will see a surrender of liberalism, other

a surrender of the hope for a conquest on liberal terms of the problem posed by the inevitable

appearance within a liberally free society of non-liberal and counter-liberal views, deeply and

sincerely held. Ferrara includes a proviso: “The norms produced through autonomous processes

[may] not enter outright conflict with the central constitutional essentials.”73 We may be left

uneasy, but Ferrara calls this “the best response to the tenuousness of consensus and the ubiquity

of dissent that political liberalism can offer.”74

VI. TAKING STOCK

38. We have noticed a number of hangups or dilemmas connected to what I called, at the

very start, “John Rawls’s seeming placement of constitutional law at the load-bearing center of

his proposed ‘liberal principle of legitimacy’”—or, in other words, that principle’s seeming

reduction of the requisite founding terms of a liberally legitimation-worthy political practice to

a body of institutionally enforced constitutional law—and a resultant cumulative assignment to

constitutional law of both regulatory and justificational burdens. Those hangups include:

(a) a self-denying exclusion from those founding terms of a commitment to the ongoing

pursuit of background justice (paras 17-19);

(b) a moral dilemma for citizens caught between convictions of substantive justice

(“human rights”) and obligations of civility, when facing questions of constitutional-

legal content and revision (paras 24-26);

(c) a dilemma of office for constitutional interpreters caught between assignments to

protect the higher law laid down by the people and to render that law in a legitimation-

worthy condition (paras 30-32);

(d) a collision of demands both for openness and closure in the presently legible

prescriptive content of the constitutional essentials (paras 33-34); and

(e) a pressure to forgo the liberal constitutional essentials, even as settled starting points

for public reasoning on matters of constitutional moment and basic justice (para 37).

39. Assuming always a place for central liberal constitutional essentials in a systemic

justification procedure, we have noticed (para 35) a difference between a treatment of them as

73 Id. at 108.

74 See id. at 106-08

27

laws that fail of their justificational purpose when appliers fail in the end to get them “right,”

and a treatment of them as settled starting points for public reasoning on matters affecting

justice at a basic-structural level. We have construed the late-coming proposal by Rawls for “an

idea of political legitimacy based on the criterion of reciprocity” (in place of the “the liberal

principle of legitimacy”?) as a move away from the former to the latter sort of conception (para

36).

We may accordingly ask: For which of our hangups does such a move provide relief. We

have already offered (para 35) to explain the move as a response to hangup (d). The move

provides relief as well for hangup (a): Rawls’s case for exclusion of a background-justice

commitment from the set of legitimation-bearing constitutional essentials depends entirely (as

I have argued) on the idea that the essentials are all and only laws that the public can expect the

Supreme Court more or less fully to police and enforce. A conception of the essentials as

mandatory considerations, in any and all of the Court’s constitutional-legal applications to which

they may pertain, seems to fit quite nicely with Sager’s proposition that sometimes a

background-justice essential will make a decisive difference in the Court’s exercise of its strictly

judicial powers and remedies in undoubtedly justiciable cases (as in the example offered above

at para 6).

40. I have not yet seen how a move to that conception can help with hangups (b), (c), and

(e). Relief from them would require, as far as I am able to see right now, one or two further

retractions from a constitution-centered legitimacy test. Political liberalism might have to give

up the call for any mandatory substantive content at all in its stipulation for “the system” that

is to bear the justificatory load—the system, that is, on the perceived legitimation-worthiness

of which will depend the moral justification for a reciprocity of demands for respectful

submission to laws issuing from it. Or political liberalism might have to give up the entire idea

of a publicly agreed and publicly certifiable legitimacy test, which citizens can cite to one

another as justification for reciprocal expectations of compliance with law;75 thus leaving

legitimacy as (i) a question that each faces and decides for himself or herself, by considering

from time to time whether the country’s overall political practice (which might or might not

include a body of substantive constitutional law) meets the demand of civility among citizens

sharing the perceptions of political reasonableness I sketched above in para 12; (ii) a

responsibility, then, for each to shoulder each and every time she deploys a modicum of power

on one or the other side of a fairly disputed matter of basic-structural justice; and (iii) a

75 See above paras 11, 15.

28

contingency, then, of a due and evident shouldering of that responsibility by a critical mass of

citizens and officials. I do not here take up the question of whether those are concessions that

a political liberalism can make.

29